Lab 2. Connect the Product API to a MySQL Database

Transcription

1 Lab 2 Connect the Product API to a MySQL Database

2 Overview In this lab, we will implement the generated subflows by connecting to a MySQL database. Step 1: Implement Get Products Flow In this step, you will implement the get:/product flow. To get the products from the database, you wiill need to: Retrieve the records from the database. Transform the records to json format. The following instructions will guide you through the process: 1. Locate the get product flow. NOTE DOUBLE CHECK: Make sure you are working in the get:\product:api-config flow 2. Now you will add the Database connector to the flow. You will see that the database connector is not in the palette. You need to add that module. 3. Go to the Mule Palette 4. Click Add Module 5. Select Database Connector. 6. Drag and Drop Icon to the light blue panel. 1

3 Going back to the Palette, you now will see the Database connector. 7. Click on the database connector, and drag and drop the Select icon before the transformation. NOTE DOUBLE CHECK: Drop the Database Connector in the Process section, not the Source section. 8. Double-Click on the Database icon and a configuration panel will be shown. 9. Change its name to Get Products. A best practice in Mule is to parameterize connectivity information. To configure the database 2

4 connector we will leverage this best practice. 10. Create a folder called config in src/main/resources 11. Create a file called configuration.yaml in the src/main/resources/config 12. Insert the following text into configuration.yaml file. mysql: host: "services.mythicalcorp.com" port: "3306" user: "product" password: "Mule1379" database: "products_test" N.B. The password can be encrypted using the Mule Credential Vault. 13. Save the configuration.yaml file 14. Go back to the api.xml file. Press the Global Elements tab. 3

5 15. Press the Create button and search for Configuration Properties 16. Press OK. 17. in the file text box write config/configuration.yaml. You can also browse to the file. 4

6 18. Now that you have your properties defined, let s add a new connector configuration. Go back to the database connector you dropped, click to open properties and click on the button on the Connector Configuration. 19. In Name write Products_Database_Configuration 20. In Connection select MySQL Connection. When you select the database connection it asks you to add the JDBC Driver. 21. Press Add dependency button. A new window will open. 22. Complete with the following data: Group id: mysql Artifact id: mysql-connector-java Version:

7 23. Click Finish. 24. For our database configuration specify the following values for the properties: Host: ${mysql.host} Port: ${mysql.port} User: ${mysql.user} Password: ${mysql.password} Database: ${mysql.database} 6

8 NOTE Mule resolves the ${key} to the configuration properties that we put in the mule-app.properties file by default without any extra configuration needed. 25. Click Test Connection to verify the connectivity 26. Click OK 27. Click OK again to close the MySQL Configuration. Now that we have the Database configured, we will add the query. 7

9 SELECT p.id, p.name, p.description, p.product_number, p.manufactured, p.colors, p.categories, p.stock, p.safety_stock_level, p.standard_cost, p.list_price, p.size, p.size_unit_measure_code, p.weight, p.weight_unit_measure_code, p.days_to_manufacture, p.images, p.modified_date, p.created_date FROM product p LIMIT 10 NOTE We are limiting the query to retrieve only up to 10 items with the "LIMIT 10" parameter. NOTE Many customers use their favorite query tool (SQL Query Analyzer, TOAD, DBVisualizer, ) to craft the query they want and then paste it into textbox in Studio. Now we need to transform the records from the database to JSON format. 28. Double-Click on the Transform Message icon. You will find the fields that come from the query on the left, and the fields that the API will return on the right. You can graphically map fields by dragging between fields. This Transform Message component uses MuleSoft s universal DataWeave transformation language for transforming data from what format to another. There is no need to use xpath/xslt for XML, code for JSON, code for CSV, etc. Use DataWeave for all transformations. Dataweave is a simple, powerful tool to query and transform data inside of Mule. 8

10 29. Remove the actual mapping and copy the mapping below into the text view of the dataweave transform. %dw 2.0 output application/json --- payload map (product, index) -> { id: product.id, categories: (product.categories default "") splitby ",", colors: (product.colors default "") splitby ",", createddate: product.created_date as String {format: "yyyy-mm-dd"}, modifieddate: product.modified_date as String {format: "yyyy-mm-dd"}, safetystocklevel: product.safety_stock_level as Number, stock: product.stock as Number, daystomanufacture: product.days_to_manufacture, name: product.name, description: product.description, images: (product.images default "") splitby ",", listprice: product.list_price, manufactured: product.manufactured, productnumber: product.product_number, size: product.size, sizeunitmeasurecode: product.size_unit_measure_code, standardcost: product.standard_cost, weightunitmeasurecode: product.weight_unit_measure_code, weight: product.weight } NOTE Since colours, categories and images are arrays and in the database they are saved as comma-separated Strings, we added to add a function called Split By to the mapping that splits the single record into a list of them separated by ",". There are a number of functions available within Dataweave to support more complex mapping requirements. See Dataweave documentation for more information. 30. This concludes this step. Click save in the transform message window. 9

11 31. Now we are ready to test our API implementation. Follow the same steps from Lab 1. Start the application. Check the logs to make sure it is deployed successfully. Test the Get Products using the GET method on the Product using the console, or Postman. You may need to set Content-Type and Accept headers to application/json. Ask your instructor for assistance if you receive error responses. 32. Now you get the response from the API, the product list is being retrieved from the MySQL database and you can check that the information is different from the one you saw in Lab 1. (Example result. Results may differ.) Now that you have confirmed this flow is working, move onto the next implementation. If you have any problems, call over the instructor before continuing. Step 2: Implement Get Specific Product Flow In this step, you are going to implement the get:/product/{id} flow. To get the product from the database, you must: 10

12 Retrieve the records from the database. Check if it was found. If it was found, transform the record to json format. If it was not found, create a "Not found" response. The following instructions will guide you through the process: 1. Remove the Transform icon from the get:/product/{id} flow. To do that you just right-click on the icon and select Delete. Or select it and hit Delete key. 2. Drag & Drop a Select component from the database connector. 3. Double Click on the icon to open the configuration tab. 4. Complete with the following parameters: Display Name: Get Product Connector configuration: Select Products_Database_Configuration SQL Query Text: SELECT p.id, p.name, p.description, p.product_number, p.manufactured, p.colors, p.categories, p.stock, p.safety_stock_level, p.standard_cost, p.list_price, p.size, p.size_unit_measure_code, p.weight, p.weight_unit_measure_code, p.days_to_manufacture, p.images, p.modified_date, p.created_date FROM product p WHERE p.id = :id Input Parameters: #[id: attributes.uriparams.id] NOTE Here we are using a parametrized query. You define the variable in the query and then you complete with the parameters on the other text box. We are going to validate if the product was found. In case it doesn t exist we are going to throw an exception. To do that we need to add the Validations Module 11

13 5. Go to the Mule Palette and add the Validation Module. 6. Select the Validation connector and drag & drop is not empty collection. The Flow should look similar to this: This validator component is going to throw an Exception of type VALIDATION:EMPTY_COLLECTION What we want to do is to catch that Exception and send a 404 (Not Found) response. 7. To do that go to the Error Handling that is defined in the api-main flow. 8. Find the component On Error Propagate type: APIKIT:NOT_FOUND 12

14 9. Click on it. You will see the configuration properties. 10. Click on the Type textbox. A checkbox list will open with all the possible errors that can be catched. 11. Select VALIDATION:EMPTY_COLLECTION 12. Press Save icon on the top right side of the panel. Now we are going to transform the response. 13. Add a Transform component at the end of the flow. copy this script into the text view of the DataWeave transform: 13

15 %dw 2.0 output application/json var product = payload[0] --- { id: product.id, name: product.name, description: product.description, manufactured: product.manufactured, productnumber: product.product_number, colors: (product.colors default "") splitby ",", categories:(product.categories default "") splitby ",", safetystocklevel: product.safety_socket_level, standardcost: (product.standard_cost default "0.0") as String {format: "##.##"} as Number, listprice: (product.list_price default "0.0") as String {format: "##.##"} as Number, stock: product.stock, safetystocklevel: product.safety_stock_level, daystomanufacture: product.days_to_manufacture, size: product.size, sizeunitmeasurecode: product.size_unit_measure_code, weight: product.weight, weightunitmeasurecode: product.weight_unit_measure_code, daystomanufacture: product.days_to_manufacture, images: (product.images default "") splitby ",", modifieddate: (product.modified_date default "") as Date {format: "yyyy-mmdd"}, createddate: (product.created_date default "") as Date {format: "yyyy-mm-dd"} } 14. Transform Message will transform the DB structure to the JSON format defined for the response. Database selects always return a list of rows, so one method to obtain a single result is to take the first one. 14

16 Finally we are going to add a Custom Business Event to register the product found. 15. Go to the Palette and search for Custom Business 16. Setup the Custom Business Event parameters to record the queried product s information for future business analysis: Display Name: Found Product Detail Event Event Name: Found Product Detail Event In the Key Performance Indicators table, add these two attributes: Name Product Name Product ID Expression / Value #[payload.name] #[payload.id] 17. Now you can run the API as in the preceding step, and check calling GET on "product/{id}" using an existing and a non-existing ID. 18. Stop the Mule Runtime using the console view after the test. Now that you have confirmed this flow is working, move onto the next implementation. If you have any problems, call over the instructor before continuing. Step 3: Implement Create Product Flow In this step, you will implement the post:/product flow. To create the product in the database, you must: 15

17 Insert the product in the DB Return the product plus the id as a response. The following instructions will guide you through the process: 1. Remove the Transform component from the post:\product:application\json:api-config flow. To do that you just right-click on it and select Delete or select it and hit the Delete key. 2. Using the Palette, drag & drop the necessary components to create a flow like this: 3. Start by configuring the Transform component. 4. Complete the Transformation with the following values: %dw 2.0 output application/java fun getmanufacturedcode(value) = if (value == true) 1 else { categories: (payload.categories default []) joinby ",", colors: (payload.colors default []) joinby ",", daystomanufacture: payload.daystomanufacture, description: payload.description, images: (payload.images default []) joinby ",", listprice: payload.listprice, manufactured: getmanufacturedcode(payload.manufactured), name: payload.name, productnumber: payload.productnumber, safetystocklevel: payload.safetystocklevel, size: payload.size, sizeunitmeasurecode: payload.sizeunitmeasurecode, standardcost: payload.standardcost, stock: payload.stock, weight: payload.weight, weightunitmeasurecode: payload.weightunitmeasurecode } NOTE Look that in Dataweave you can declare functions and use if else structures. This is use to transform the manufactured value from boolean to integer. 16

18 5. We will add a second mapping in the same Transform to save the original payload so it may be used for the response. Click on the "Add new target" button. 6. In the popup ensure the output type is set to "Variable" and set the variable name to "originalpayload". Click OK. 7. In this second mapping will use the following simple script to save the full input payload. Verify you see "FlowVar - originalpayload" in the Output. You can switch between script using the down arrow. %dw 2.0 output application/java --- payload 8. Configuring the Database. Set the parameters as follows to save the product into the database: Display Name: Insert Product Connector configuration: Products_Database_Configuration SQL Query Text: INSERT INTO product(name, description, product_number, manufactured, colors, categories, stock, safety_stock_level, standard_cost, list_price, size, size_unit_measure_code, weight, weight_unit_measure_code, days_to_manufacture, images, modified_date, created_date) VALUES(:name, :description, :product_number, :manufactured, :colors, :categories, :stock, :safety_stock_level, :standard_cost, :list_price, :size, :size_unit_measure_code, :weight, :weight_unit_measure_code, :days_to_manufacture, :images,curdate(), CURDATE() ); Input Parameters: 17

19 #[{name: payload.name, description: payload.description, product_number: payload.productnumber, manufactured: payload.manufactured, colors: payload.colors, categories: payload.categories, stock: payload.stock, safety_stock_level: payload.safetystocklevel, standard_cost: payload.standardcost, list_price: payload.listprice, size: payload.size, size_unit_measure_code: payload.sizeunitmeasurecode, weight: payload.weight, weight_unit_measure_code: payload.weightunitmeasurecode, days_to_manufacture: payload.daystomanufacture, images: payload.images}] 9. As this operation creates a new product, you will need to obtain the newly generated id. How this is implemented can vary with different database engines' drivers. 10. In the case of MySQL, go to "Advanced" in Database configuration. 11. Find Auto Generated Keys section. 12. In the *Auto generated keys columns names" select Edit inline. a. Add value id in the table below. The configuration will be like this: Now we are going to configure the response Transform Message. In this step we need to merge the original request payload with the returned ID from the database. The next script will do the trick. We are getting the original request payload and adding a field called "id". 18

20 13. Add the following transformation: %dw 2.0 output application/json --- vars.originalpayload ++ id: payload.generatedkeys.generated_key 14. Finally let s configure the Custom Business Event. Set the parameters to record the created product s information for future business analysis: a. Display Name: New Product Created Event b. Event Name: New Product Created Event c. In the Key Performance Indicators table, add these two attributes: Name Product ID Product Name Expression / Value #[payload.id] #[payload.name] 15. Now you can run the API as in the preceding steps, and check it is working by calling POST on "products". CAUTION The productnumber must be unique, you need to change the productnumber with a random value. If you have problems with this, ask your instructor to provide a value by querying the database directly to check for existing records Stop the Mule Runtime using the console view after the test. Now that you have confirmed this flow is working, move onto the next implementation. If you have any problems, call over the instructor before continuing. 19

21 Step 4: Implement Delete Product Flow In this step, we are going to implement the delete:\product{id} flow. To delete the product in the database, we just need to call the database. The following instructions will guide you through the process: 1. Remove the components from the delete:\product{id} flow. To do that you just right-click on each component and select Delete or select a component and hit the Delete key. 2. Using the Palette, drag & drop the needed components to create a flow like this: 3. Configure the Database component as follows: Display Name: Delete Product Connector configuration: Products_Database_Configuration SQL Query Text: delete from product where id=:id Input Parameters: #[id:attributes.uriparams.id] Http Response 204 means No Content. So we are going to send a Null Payload. So configure the Set Payload with the response. 4. Click on the Transform Component. As we did before, let s create a variable with the http status 5. Click on the "Add output" button 6. In the popup, set the variable name as "httpstatus" and click OK. 7. Copy and paste the following text: 20

22 %dw 2.0 output application/java Switch back to the payload. Let s remove the payload from the transformation. 9. Press the trash icon. This will remove the payload. 10. Now you can run the API as in the preceding steps, and check it is working by calling DELETE on "products/{id}" for an id you know exists (such as the one created in the previous step). Note that if you have more time you can extend the implementation as in the GET /products/{id} flow to respond with a not found response if the record does not exist. Stop the Mule Runtime using the console view after the test. Now that you have confirmed this flow is working, move onto the next implementation. If you have any problems, call over the instructor before continuing. Step 5: Implement Update Product Flow In this step, we are going to implement the put:\product{id} flow. To update the product in the database, we just need to call the database. The following instructions will guide you through the process: 1. Remove the components from the put:\product{id} flow. To do that you just right-click on each and select Delete or select it and hit Delete key. 2. Using the Palette, drag & drop the needed components to create a flow like this: 21

23 3. Configure the Transform Component with the following script. Note the boolean to number transformation in the manufactured field to match database format. %dw 2.0 output application/java fun getmanufacturedcode(value) = if (value == true) 1 else { categories: (payload.categories default []) joinby ",", colors: (payload.colors default []) joinby ",", daystomanufacture: payload.daystomanufacture, description: payload.description, images: (payload.images default []) joinby ",", listprice: payload.listprice, manufactured: getmanufacturedcode(payload.manufactured), name: payload.name, productnumber: payload.productnumber, safetystocklevel: payload.safetystocklevel, size: payload.size, sizeunitmeasurecode: payload.sizeunitmeasurecode, standardcost: payload.standardcost, stock: payload.stock, weight: payload.weight, weightunitmeasurecode: payload.weightunitmeasurecode } 4. Configure the Database with this parameters: Display Name: Update Product Connector configuration: Products_Database_Configuration SQL Query Test: update product set name = :name, description = :description, product_number = :product_number, manufactured = :manufactured, colors = :colors, categories= :categories, stock = :stock, safety_stock_level = :safety_stock_level, standard_cost = :standard_cost, list_price = :list_price, size = :size, size_unit_measure_code = :size_unit_measure_code, weight = :weight, weight_unit_measure_code = :weight_unit_measure_code, days_to_manufacture = :days_to_manufacture, images = :images, modified_date = CURDATE() where id = :id Input Parameters: 22

24 #[{name: payload.name, description: payload.description, product_number: payload.productnumber, manufactured: payload.manufactured, colors: payload.colors, categories: payload.categories, stock: payload.stock, safety_stock_level: payload.safetystocklevel, standard_cost: payload.standardcost, list_price: payload.listprice, size: payload.size, size_unit_measure_code: payload.sizeunitmeasurecode, weight: payload.weight, weight_unit_measure_code: payload.weightunitmeasurecode, days_to_manufacture: payload.daystomanufacture, images: payload.images, id: attributes.uriparams.id}] Like we did in the last Step, we are going to return On the last Transformation component. Click on the "Add output" button 6. In the popup, set the variable name as "httpstatus" and click OK. 7. Copy and paste the following text: %dw 2.0 output application/java Switch back to the payload. Let s remove the payload from the transformation. 9. Press the trash icon. This will remove the payload. 10. Now you can run the API as in the preceding steps, and check it is working by calling PUT on "products/{id}" for an id you know exists. Note that if you have more time you can extend the implementation as in the GET /products/{id} flow to respond with a not found response if the record does not exist, however note that in the restful style, it is usual to invoke a create operation on the resource if it does not already exist. Stop the Mule Runtime using the console view after the test. If you have any problems, call over the instructor before continuing. Now that we have a working implementation, in the next lab we will test it end to end and add unit tests. Summary In this module we implemented the System API against the MySQL database. We ran and tested the API on the local Mule Runtime which comes embedded in Anypoint Studio. Overview You completed the following steps: Implement Get Products Flow 23

25 Implement Get Product by ID Flow Implement Create Product Flow Implement Delete Product Flow Implement Update Product Flow You were introduced to the use of Dataweave, a powerful tool to transform messages from a database record to json format in this case. You also saw how simple accessing a database can be using the Database Connector. It is important to note that the implementation of an API is a real Mule application. Building integration applications and exposing APIs do not require additional knowledge, making it easy for Mule developers to work on both integrations and APIs. See the Dataweave documentation for more information. See the Database Connecter documentation for more information. Please proceed to Lab 3: Run and Debug the API Take me to the TOP 24